LIMNOLOGY AND

OCEANOGRAPHY

April 1959

VOLUME IV

NUMBER 2

Stratigraphic Distribution of Amino Acids in Peats from Cedar Creek Bog, Minnesota, and Dismal Swamp, Virginia’

F. M. SWAIN, A. BLUMENTALS, AND R. MILLERS

University of Minnesota, Minneapolis

ABSTRACT

The concentration of several alpha-amino acids obtained by acid hydrolysis at various levels in Cedar Creek Bog and Dismal Swamp has been determined. Seven amino acids are consistently present in the natural accumulations: glycine, aspartic acid, glutamic acid, threonine, alanine, valine, and leucine. Alanine occurs in the highest concentrations, attaining about 22 pmol/g in the samples investigated. Free amino acids were absent or very rare in both the bog waters and waters squeezed from the peats. Attempts to detect proteins in these substances by means of paper electrophoresis were unsuccessful. Fibrous proteins may be present, however.

In both peat deposits there is a greater concentration of amino acids at depths of a few feet than in the uppermost peat layers. Hydrolysis of a humic acid preparation from Minnesota peat yielded the same suite of amino acids given above. It is therefore suggested that the amino acids in our peat hydrolyzates mainly occurred adsorbed by, or linked to, the humic acid micelles rather than as proteinaceous compounds.

The distribution of amino acids in Cedar Creek Bog peat suggests two periods of trophi- cation, one beginning approximately 12,000 years ago in Mankato time and ending with the so-called thermal maximum perhaps as late as 6,000 years ago, the other beginning after the thermal maximum and continuing to the present. Evidence of more than one cycle of peat development in the Lake Drummond area of Dismal Swamp is lacking.

INTRODUCTlON

The nitrogen content of peat deposits reaches 4 % or more but probably averages less than 1% (Waksman 1938, p. 277). In Cedar Creek Bog, nitrogen content is 2.3 % at 4-5 feet, 2.8% at 7-8 feet, 3.4% at 9-10 feet, and less than 1% at most deeper levels (Swain and Prokopovich 1954, p. 1186). The nitrogen generally is considered to occur in the form of proteinaceous compounds that have formed as cell constituents of micro-

1 The work has been supported by grants from the National Science Foundation and the Gradu- ate School, University of Minnesota. The writers are indebted to K. E. Dickinson and Harvey Meyer for assistance in collecting the Cedar Creek Bog samples. H. E. Wright, Jr., and J. G. Pala- cas kindly read the manuscript and offered helpful suggestions. Prof. H. Kleerekoper reviewed the manuscript and generously supplied a copy of his book on lake sediments of southern Ontario.

organisms. The latter have grown in the subsurface portions of the peat bog from the decomposition products of the accumulating higher plants (Waksman 1938, p. 115). The downward increase in total nitrogen, typi- cally found in low-moor peats (Waksman and Stevens 1928) is offered in support of this concept. The information resulting from the present study suggests, on the other hand, that the nitrogenous materials in the two lowmoor bogs we studied are associated primarily with humic acids rather than with proteins in the strict sense, al- though existence of peptide linkages in the material cannot be ruled out.

The vertical distribution of amino acids in the upper 2-4 meters of lake sediments in southern Ontario was studied by Kleere- koper (1957, p. 124). He found that the amino acids showed little change with depth

119

120 SWAIN, 13LUMENTRLS, ANI) MILLERS

in the scdimcnts of some of the lakes, an in- ing alnninc, arginine, aspartic acid, cystine, crease in depth in others, and a decrease in glutamic acid, glycine, hydroxyprolinc, lcu- still others. The changes apparently arc rc- cinc and isoleucine, lysine, methionine, pro- lated to variations in the sand, clay, and line, serine, tourine, threonine, and valine. perhaps marl content of the lake sediments, An excess of nitrogen not accounted for by rather than to migrations within the scdi- the amino acids is thought to occur in mcnts as here suggested for Cedar Creek lignin-nitrogen complexes, such as humic Bog and Dismal - Swamp peats. Klcere- acids. koper was successful in isolating a rather The distribution of humic substances in large number of amino acids by sulfuric acid several European lakes was thought by hydrolysis from the lake sediments, includ- Ztillig (195G, p. 109-111) to result not only

TABLE 1. Vertical distribulion of amino acids in hydrolyxates of Ccdur Creek peat, Minnesota

Amino acids (parts per 10,000) wt. of

Bog wet

Zones YYe

Mois ture

% DcI:th Type of sample

-- .- 2-3 Cogropclic peat IIf 17 5-6 P&y copropcl @I) 21 8-9 Sapropclic copropel PC 20

11-12 Sapropelic copropcl PC 21.5 14-15 Copropclic marl MC 23 t7-18 Copropelic marl v 1 c 29 20-21 Copropclic marl M, 28 23-24 Copropclia marl l&r, 23 26-27 Dark copropcl PC 12.5

29-30 Copropolic marl MC 30

32-33 Copropclic marl M 37 35-36 Sideritic marl Myid 24 37-38 Sidcritic sand rS 42.9 41-42 Sand rs 28

Bison Ilark, humus im- 10 rib pregnatcd (tlry)

-.- _.____-

-

z WI N% t’hrc- mine

--

0.20 2.83 1.77 1.74 0.80 0.67 0.02 0.24

-

0.14 0.10 0.39 0.12

-

.030

-..

Alanine Va- line

--- _

2.94 21.92 11.25 19.77 0.49

- -. -

0.27 -

0.03 0.63 1.68

-

3.098n 3.0864pa

1.12 2.83 5.82 3.02 1.67 0.15 -

0.14 0.17 0.25 0.02 0.19 0.21

-

1.198 0

--- -

Leu- cine

2.06 0.65 9.33 48.07 7.82 31.71 7.37 38.29 1.48 6.73 0.15 3.04

- 0.13 0.20 1.14 0.40 1.31 0.45 1.12 0.01 0.25

-- 2.13 0.31 2.55

- -

1.288 4.410

ills G1u.t. acid

__- -~

0.94 0.60 1.95 5.68 O.GO 3.55 1.16 3.40 0.28 1.58 0.28 1.37 0.07 0.03 0.03 0.48 0.19 - 0.01 0.25 0.03 0.02 0.05 0.80 0.005 0.24

- -

2.996 0.254

Ill

Xycine

1.7 3.53 0.90 1.74 0.43 0.42 0.01 0.05 0.28 0.02 0.04 0.07 0.007

-

0.236* 0.224h

89 90 87 68 01 60 GO 60 70 GO 60 70 GO 35

47.23f 47.59 48.01

9.78 15.13

- -- - - -

8.47 -- - -

2.31 2.77 3.36 0.90 0.75

-- - _-

- -

0.9 _- -- -

* probably includes serine and aspartic acid t C and N determinations were made previously on cores collected near “b” on lcigure 1. Depths arc, respcctivcly: 4-5,7-S, 9-10,

11-12, 12-14, and 35-36 feet. . Iv = lysine

.__ /“- a = alanino pa = phenylalanine ?J = histidine

WC’

TABLE 2. Vwtical distribulion of amino acids (in parts per 10,000) in acid h,ydrolyzales of Dismal Swamp peat, Virginia -.--

De th pt

----

l-2

2-3

3-4

4-5

5-6

6-7

daninc

2.49

.oo

1.81

2.88

4.91

1.43

-.

. P

- -

1

-.

_- --

Valine

0.99

.OOE

2.16

0.59

0.86

0.37

__- -

--

Bog ones

.-

C

_ -

--

ilycinc

0.32

-

1.46

0.44

0.53

0.14

-

r Glu- t;::

--

0.16

I’hrco- nine

0.08

- -

7.16 3.54

0.54 0.27

0.91

0.37

- --

0.45

0.18

wt. of wet*

ample (9)

28

27

40

60

,eucme 1$2J- 2: Lysine

0.08

-

0.37

0.11

1 Type of sample

Reddish brown woody pest

Reddish brown woody peat

Reddish brown woody peat

Dark brown copropeljc fibrous peat

Dark brown copropclic fibrous peat

Sandy, dark fibrous copropcl

S --

5 .005 - .015

1.51 0.50 28.52

0.95 0.82 6.60

1.74 0.18 9.71

0.66 0.11 3.30

__--_-~-- - - - --.- -- ----. * Moisture content approximately 90%.

STl~ATlGlkWHIC DISTRIBUTION OF AMINO RClDS IN l’l!XTS 121

from autochthonous production but from allochthonous introduction of organic matc- rial. An increase of organic nitrogen 5-10 cm beneath the surface in the bottom scdi- mcnts of eutrophic Ztirichscc was attributed to allogcnic material, The prcscnt writers rccognizc this possibility but bclicvc that because trophic&ion has been generally pro- gressive in Cedar Creek Bog and Dismal Swamp, the humic substances and associated amino acids, whether of authigenic or allo- genie nature, should typically increase up- ward to the surface. That this is not the case is bclievcd to have resulted from down- ward concentration of soluble humic ma- terial rather than from increased allogenic influx of humus at some time in the past.

METHODS

Wet samples of the peat and underlying marl were hydrolyzed with 6N hydrochloric acid for 22 hours under rcflux. The hy- drolyzates were reduced to dryness and taken up in 5 ml of isopropyl alcohol. Ap- proximately 1 microlitcr of the concentrates of each sample was placed on sheets of Whatman No. 1 filter paper together with a mixture of known amino acids; solvent, butanol : acetic acid : water (40 : IO : 50) ; chro- matograms stained by dipping in ninhy- drin in acetone. The devclopcd chromato- grams wcrc then scanned with a photometer, and the concentrations in the wet sediment were determined as shown in Tables 1 and 2, using the area x density method (Block et al. 1955, p, 72). The approximate dry weight values can be determined from the moisture content, If an average weight of a mole of the amino acids studied here bc taken as 100 g, the values in Tables 1 and 2 can bc read in ,umol/g. Two-dimensional chroma- tograms were run to verify the identity of the individual amino acids (Fig. 2B). A com- posite chromatogram of the suite of Cedar Creek Bog samples is shown in Figure 2A.

AMINO ACID DISTRIBUTION IN

CEDAR CREEK BOG

The vcgctational zones and collcc ting localities at Ceda.r Creek Bog are shown in Figure 1. In the upper five feet of the bog thcrc is a downward increase in all the amino acids from as little as 2-fold in as-

partic acid to as much as IO-fold in alaninc. The upper 4-5 feet of the sediments of Cedar Creek Bog arc characterized by a predom- inance of forest and Sphagnum peat; below lies scdgc peat and copropel (gyttja). Part of the downward increase in amino acids is believed to bc due to the originally greater protein content in the sedge peat and copro- pel than in the forest peat. Additional downward concentration of amino acids may bc accomplished by means of aqueous solu- tions of “humic acid” to which the amino acids are attached. Because the copropelic material is relatively impervious, the bog wa- ters from above probably do not circulate across thcsc layers to any extent, The amino acids brought into the upper part of the copropcl by the moving bog waters probably are conccntratcd and held there.

Water humus (Waksman 1938, p. 290) is soluble or occurs as gel suspensions in the waters of bogs and especially of dystrophic lakes. A special study of some of this ma- terial was made by Shapiro (1957), who found thcsc “humolimnic” acids to be of phenolic or cnolic nature with an empirical formula of (C&l&027N). FIc bclicved the nitrogen to bc an impurity. The present writers suggest that amino acids may be attuchcd to these humic materials perhaps by means of an amine bridge as postulated by Thielc and Kcttner (1953)2 and thus subject

2 Thcsc authors proposed that humic acid formed from quinones, such as p-bcnzoquinonc probably by bacterial decomposition :

0 I

2n 1\ 9 + Ill-12 0 -3 \

II 0

p-bcnzoquinonc

OII

n + (CsILOJn

\ humic acid

OH OII 0

hydroquininc hydroxyquinone

122 SWAIN, BLUMENTALS, AND MILLEXS

to transportation vcrticully or laterally within the bog.

Samples of the surface bog water and of water squcczcd from the peat samples were concentrated and tested for proteins by means of paper electrophoresis, and for amino acids by paper chromatography. Water soluble proteins apparently arc ab- sent, and free amino acids are lacking or nearly so in these waters. When the peat samples were extracted with cold dilute HCl, differences in amino acid content of the extracts differed from that of the hydrolyzatc only in total amount. The data suggest that free amino acids arc rare in Cedar Creek Bog. The amino acids freed by decomposi-

l?ootnote 2 continued. The humic acid benzenoid system may form by polymerizing the hydroxyqumone, as follows :

OH OH I 0 I 0 /u + /u +

0 0

Thiclc and Kettncr visualized that nitrogen might occur in the humic acid molecule as a bridge substance of the nature of oxazine:

This mechanism may also provide a means of incorporation of amino acids in the humic acid micclle.

tion of the bog vegetation are either: (1) immediately utilized by other bog organ- isms, or (2) are bound to the humic acid benzcnoid structures that are forming at the same time

Supporting evidence for this conclusion is the presence of amino acids in, or asso- ciated with, humic acid preparations as noted by several observers (Okuda and Hori 1954, 1956; Hayashi 1956; Bremncr 1955; Blumcntals and Swain 1956). At the present time it is not known whether the amino acids arc simply adsorbed to the humic micelle or linked to it by -CHZ-,

H -N--, -CO---, -S-, or other group. The near neutrality of the bog waters would favor the transportation of the humic acid micellcs and their associated amino acids laterally and downward, thus helping to ac- count for the observed distribution in the upper lo-12 feet. The downward moving humic acids would be prccipitatcd in part by flocculation on contact with the underlying marl, as well as by the imperviousness of the copropel.

Beneath the copropcl there is a sharp de- crease in amino acid content of the under- lying marl that results not only from the lower total organic content of the marl, but also from the lack of downward migration of the high molecular-weight substance through the copropel. There is a general downward decrease in the individual amino acids between depths of 14 and 26 feet; in fact alanine is absent at several levels. This is very likely the result of downward con- centration in the relatively porous marl.

The increase of several of the amino acids in the dark copropcl layer at 26-27 feet may be the result both of original high protein composition and of downward concentra- tion. This irnpervious layer probably im- pcdcs further concentration downward of amino acids from the overlying marl. Be- neath the dark copropcl, the generally low amino acid content is a reflection both of low total organic content and downward concent&ion. Alaninc, particularly, seems to move downward to the nearest impervious or flocculating layer; in this case sidcritic marl and sand.

TO test stability of the Cedar Creek

STRATIGRAPRIC DISTRIBUTION OF AMINO ACIDS IN PEATS 125

tration noted in Cedar Creek Bog occurs in Dismal Swamp as well, and (2) whcthcr the amounts and kinds of amino acids showed similar relations to stratigraphy in the two areas.

The samples of Dismal Swamp peat were obtained on Lake Drummond about 500 feet north of the mouth of Dismal Swamp Canal feeder. The locality is close to Osbon’s (1919, p. 48) locality D where he found 10 feet of peat having, at a depth of 4 feet, the following moisture-free composition: C, 20.66 %; volatile matter, 53.02 % ; ash, 26.32%; N, 1.84%; S, 0.31%. This is re- ferred to as “black gum” peat because of the abundance of the gum tree Nyssa BiJlora in the area. The following are also found in the black gum association (Osbon 1919, p. 43): Fraxinus caroliniana (water ash), Berchemia scandens (rattan), Gelsemium sempervirens (yellow jcssaminc), Bigonia capreolata (cross vine), l’axodium distichum (bald cypress), Acer rubrum (red maple), and Nyssa aquatica (cotton gum).

At the locality sampled there were about 7 feet of peat, the upper 4 feet consisting of reddish brown, porous, punky, woody peat. The lower 3 feet arc dark brown, copropclic, fibrous peat, very sandy in the lower 6 inches and rest upon sand. The peat waters are more acidic than those of Cedar Creek Bog ranging between 4.5 and 5.

The amino acids attain their highest con- centration in the lower part of the porous woody peat. In the hydrolyzate from 2-3 feet beneath the surface no amino acids wcrc detected. Because of the rapid growth of vegetation in this swamp, the surface layers probably contain undecomposed pro- tein or polypeptidcs plus humus. The amino acids may occur in both types of com- pounds here. The humic acids and asso- ciated amino acids appear to undergo con- centration downward to the base of the woody peat where they are precipitated. The absence of amino acids in the 2-3 foot layer is believed to be a result of the down- ward movement of the water-soluble humic acids that are continually leached from the upper layers of peat as fast as they form. The effect on amino acid distribution of the many swamp fires that have swept the area is not definitely known; perhaps there has

been merely a general loss of organic con- stituents rather than of particular constit- uents. It does not seem reasonable that reduction of amino acids in the 2-3 foot layer resulted from their destruction by heat, because the peat of this layer is not observably different from that above or below.

The age of the sands underlying the peat may bc Pliocene (Shalcr 1890, p. 315) or Pleistocene (Osbon 1919, p. SO), but what- ever their age, there is no evidence in the samples studied to suggest more than one cycle of peat development in the Lake Drummond area of Dismal swamp. The lower dark layers represent lake peat that was succeeded by the reddish brown forest peat. In other parts of Dismal Swamp thicker peat deposits have been reported, and in those places the bog development, may have been more complex.

PROFILE DEVELOPMENT IN TIIE PEAT BOGS

The distribution of the layers in the typical peat bog from the viewpoint of genesis and distribution of decomposition products can be compared to a forest or meadow soil profile. The symbols sug- gcsted here are for USC in describing the several peat layers. The active dccomposi- tion layer (110) can be likened to the A0 soil zone of decomposition; the underlying leached zone (II) in the peat is comparable to the &-AZ soil zone; and the ,G’ zone of concentration of humic acids is similar to the B soil zone. The zones in the peat bog of which Cedar Creek Bog, Minnesota, is a type example are summarized as follows : 110 zone; l?ermentation layer, characterized

by moderate content of carbohydrates, lignin, bitumens, humic substances and amino acids; late eutrophic and/or dys- trophic stage of bog.

IIf, II,, etc. zone; Bleached layer in forest peat, sedge peat, etc., indicated by sub- script; characterized by lower content of carbohydrates and amino acid, and highcl content of lignin and humic substances; eutrophic and/or dystrophic stage of bog.

&, PO, etc. zone; Accumulation layer in peat, copropel, etc., shown by subscript; char- acterizcd by high content of humic sub- stances and bitumen, increased quantities

126 SWAIN, BLUMENTALS, AND MILLERS

of amino acids associated with humic ma- terials, reduced lignin, and markedly re- duced carbohydrates, eutrophic and/or dystrophic stage of bog.

M zone; marl layers deposited during photo- synthesis of phytoplankton, representing interval of introduction of calcium bicar- bonate brought into the lake in solution from surrounding uplands; the marl is frequently copropelic and w&h siderite layer at base; early eutrophic stage of bog.

I’,, I‘,, etc. zone: sub-bog material consisting of sand, clay, etc.

T’OSSIBLE EFFECTS ON Cl” DETERMINATIONS

The effects of downward concentration of amino acids on radiocarbon age determina- tions would be to increase the apparent age of the leached zone and to decrease that of the accumulation layer. The quantities of amino acids in Cedar Creek Bog are such that the cffccts of downward migration would bc of the order of .Ol to 5 %; i.e., in the richest layer, 5-6 feet, if all the amino acids at this level had been carried from above, an error of about 2 % of the total sample would be involved. The carbon content of the sample is approximately 45 % (Swain and I’rokopovich 1954, p. 1186).

In addition, downward concentration of some other organic materials such as the humic acids also may involve errors of the order of 5 % or more, judging from prelim- inary data WC have assembled on the dis- tribution of the humic materials. The total movement of organic material may involve quantities of 1 to 10 % of the sample, thereby introducing corresponding errors in the radiocarbon dctcrminations. Objects such as bone, shell, or wood that tend to adsorb migrating substances may bc particularly prone to error in radiocarbon data. Re- moval of the humic material from the spccimcn prior to treatment for radiocarbon dating might help to avoid these possible errors.

LATE PLEISTOCENE EVENTS RECORDED

IN CEDAR CREEK BOG

The succession of amino acids and of marl and peat in Cedar Creek Bog suggest two periods of lake development. The first

stage began approximately 12,000 years ago (Wright and Rubin 1956) with the deposition of the basal sideritic marl in a relatively cold, but post-glacial, climate, the latter indicated by the pine pollen of Zone B (Flint and Decvcy 1951, p. 272). An ap- proximate correlation of the base of Cedar Creek Bog with the Two Creeks horizon of north-central U. S. and the Allcrijd horizon of Europe is suggested by Cl4 dates (Flint 1957, p. 395). After about 12 feet or more of marl had accumulated, the trophication of the lake had progressed to the stage in which dark copropel (gyttja) formed plenti- fully on the lake floor, at which time the vegetation surrounding the lake was dom- inated by oak (pollen zone C-2) (Flint & Deevey 1951, p. 272), suggesting that the post-glacial warm period had been reached. A radiocarbon date of 7988 rt 420 years from the top of pollen zone I3 was obtained (Flint & Deevey 1951, p. 272) from the marl beneath the dark copropcl.

The apparent original distribution as well as subsequent concentration in the humus of the amino acids in Cedar Creek peat and marl suggest two periods of trophication. The thermal maximum in the Cedar Creek area may have been of short duration, a few hundred years or less, because only copropel, not scdgc or forest peat, is found. Whether or not aridity was necessary to reinstate the deposition of marl that ovcr- lies the dark copropel is uncertain. Ele- vated temperature and lessening of precipi- tation probably resulted in general lowering of the groundwater level, and of the zone of weathering during the thermal maximum. Lowering of temperature and rise of ground- water level in the time subsequent to the post-glacial warm interval has rcsultcd in calcium carbonate again being leached from the glacial drift to bc carried into Cedar Creek Bog. This second cycle of trophica- tion has apparently continued until the pres- cnt time with no pronounced interruptions.

CONCLUSION

Additional work should be done on the fulvic and humic acid fractions of peat, as well as on the alkali-insoluble humin to dc- terminc the nature of the nitrogenous com-

STRATIGRAPIIIC DISTRIBUTION OF AMINO ACIDS IN PEATS 127

pounds and the precise way the amino acids appear to bc associated with them. Amino acids have been found in rocks as old as Koccne (Vallentync 1957) and in fossils as old as Devonian (Abelson 1954, 1957), wherein association with humic substances is likely. Various other bituminous rock samples as old as Precambrian have been investigated for amino acids by the writers with variable success. The mechanism by which certain amino acids arc preserved in some rock materials and not in others should be studied.

The widespread distribution of a few amino acids in bog and lake deposits is of intcrcst because several of the so-called essential amino acids arc absent from the list or occur only infrequently (isoleucine, phenylalanine, methionine, arginine, histi- dine), and because several that arc found in oil shale and in fossil shells also arc absent. The medium in which the amino acids occur may be significant, because Bison rib from a Minneapolis peat bog contained several amino acids (histidine, lysine, serine (?), and phcnylalaninc) not detected or rare in the peat.

The amino acids observed by Abelson (1957, p. 279) to occur most commonly in fossil Mercenaria from l’leistocenc, Pliocene, and Mioccnc strata arc glutamic acid, gly- cine, alaninc, isolcucinc or leucine, prolinc, and valine. Of these only prolinc was not commonly found in the peat deposits. As- partic acid which Abelson suggests is rcla- tively easily hydrolyzed to ammonia and malic acid at moderately elevated tcmpera- turcs is commonly present in the peats and may serve as a geologic thermometer in in- terpreting the past history of organic scdi- mcnts.

REFERENCES

A\BETJ~ON, P. 1.1. 1954. Organic constituents of fossils. Ann. Rept. of Dir. of Gcophys. Lab., Carncgic Inst., 97-101.

Wash., Paper No. 1235, pp.

---- 1957. Some aspects of palcobiochcm- is&y. Ann. N.Y. Acad. Sci., 69(2) : 276-285.

RRNOT,D, J. R., AND W. F. KIRBY. 1951. Radio- carbon dates. Scicncc, 113: 117.

.BLoCK, R. J., IX. 1,. T~JRRUM, AND C. ZWEIG. 1955. Paper chromatography electrophoresis.

and paper Academic Press, N. Y.

l-484.

BLTJMENTALS, A., AND F. M. SWAIN. 1956. Com- parison of amino acids obtained by acid hydrolysis of lake sediments, central Minnc- sota. Geol. Sot. Am. Bull., 67: 1673.

BREMNER, J. M. 1955. Nitrogen dist,ribution and amino acid composition of fractions of a humic acid from a chcrnozem soil. Zeit Pflanzencrnahr. Diing. Bodcnk., 71: 63-66 (in English).

FLINT, It. F. 1957. Glacial and Pleistocene Geology. John Wiley and Sons, New York. vi-xiii, l-553.

I~JNT, It. F., AND E. S. T)EEVEY, Ja. 1951. Radiocarbon dating of late-Plcistoccnc events. Am. J. Sci., 249: 257-300.

HAYASIII, T. 1956. Components of soil humic acid III. Nitrogenous constituents of A. Type humic acids. Sci. Soil Manure, Japan, 26: 371-375.

KLEEREKOPER, 11. 1957. Unc etude limnol- ogiquc dc la chimic dcs sediments de fond dcs lacs de 1’Ontario meridional Canada. Uit- gcvcrig Excelsior, ‘s-Gravenhage. 205 pp.

OKUDA, A., AND S. 11~~1. 1954. Chromato- graphic investigation of amino acids in humic acid and alkaline alcohol lignin. Mcm. Jtcs. Inst. Food Sci., Kyoto Univ., No. 7: l-5.

--- . 1956. Idcntificution of amino acids in humic acid. Sci. Soil lManurc, Japan, 26: 346-348.

OSBON, C. C. 1919. Peat in the Dismal Swamp, Virginia and North Carolina. U. S. Geol. Survey Bull., 711C: 41-59.

STrRLWR, N. S. 1890. Fresh-water morasses of the United States. U. S. Gcol. Surv. Ann. Rept., 10: 261-339.

STIAPIRO, J. 1957. Chemical and biological studies on the yellow organic acids of lake water. Limnol. Oceanogr., 11: 161-179.

SWAIN, F. M., AND N. hOKOPOVICJ1. 1954. Stratigraphic distribution of lipoid sub- stances in Cedar Creek Bog, Minnesota. Gcol. Sot. Am. Bull., 66: 1183-1198.

TTIIET,IX, II., AND Ii. KETTNER. 1953. Humin- sauren. Kolloid-Zeit., 130: 131-160.

VALLENTYNE, J. R. 1957. The molecular naturo of organic matter in lakes and oceans with lesser rcfcrcnce to sewage and tcrrcstial soils. J. Fish. Rcs. Bd. Canada, 14: 33-82.

WAKSMAN, S. A. 1938. Humus. Origin, chcmi- cal composition, and importance in nature. Williams and Wilkins Co., l3al timoro.

WAKSMAN, S. A., 526 pp.

AND K It STEVENS. * . 1928. Contribution to the chemical composition of peat. I. Chemical nature of organic com- plexes in peat and methods of analysis. Soil Sci., 26: 113-137, 239-252.

WRIGHT, TT. E. JR., AND M. RUBIN. 1956. Radio- carbon dates of Mankato Drift in Minnesota. Scicncc, 124: 625626.

Z-~~T,LIG, II. 1956. Sedimcnto ala husdruck des Zustandes tines Gewasscrs. Schweiz . Zei t. IIydrol., 18: 5--143,